I think your prediction makes sense. There needs to be a lot of guardrails and you need a fully integrated platform/stack on the backend to protect the user from doing dangerous things because the AI ain't gonna tell them when something is dangerous or doesn't perform at the scale they need

We should be able to just deal with pure UX issues/tradeoffs; we shouldn't be running into technical issues.

The more feature-dense your code is, the less unnecessary hurdles and complexity, the more reliable your software is and the easier to extend and maintain. The more code can fit into the LLM's context window, the more complex your application can become whilst remaining reliable.

I built my own framework which is declarative HTML web components with a serverless backend (similar to Firebase but optimized for AI so that the AI can interact with the control panel, impersonate users and test access controls by itself) and it's so easy to create new features. It's just pure UX work. Just telling the AI what the problems are. Once fixed, it's fixed for good. By contrast, with a standard framework, I have to spend a lot of time debugging technical problems and keep running into regressions and it takes way longer to fix anything and requires coding knowledge.

LLMs work really well with custom frameworks provided that they're well documented via agent skills.

React doesn't provide optimal abstractions for safe and efficient LLM coding IMO; when building applications, React still makes you duplicate a lot of work and create a lot of custom components. A lot of the wiring has to be re-created over and over. If you really think about the essence of a web UI; it's just sections with lists of records, single records (loaded based on ID), single fields of specific records, forms... And each one requires access control rules which should be enforced by the backend; ideally in a declarative way so that it's easy to audit. IMO, the backend access control rules should be exhaustively auditable so that the AI can impersonate all the roles and access all possible resources in all possible states to verify that access control is enforced correctly in all possible situations. Anything less than that is not suitable for vibe-coding IMO. A user who doesn't even know what access control means cannot possibly build a secure application.

Last time I tried my hand at whitehat hacking on HackerOne, it took me 30 minutes to find a major system crash/DoS vulnerability in a major platform. The company acknowledged that the issue was real but denied me the bounty payment because they said I would have to 'prove' that it leads to catastrophic failure. I had already done so in the sense that you could reliably infer it from the data I had provided, but it seemed like they were baiting me into committing a felony (DoS attack) to prove my point, which I wasn't prepared to do but I'm sure I could have done cheaply. So yeah, whitehat hacking seems to be a waste of time. Most software today is incredibly insecure.

Also, I don't think it's useful because AI involvement in the work is a spectrum, not a binary true or false. If I ask Claude to go over my code and fix up typos and clean up and it starts adding comments with hidden watermarks everywhere it makes the tiniest adjustment, it's essentially appropriating my code. Its contribution may be trivial and superficial.

When I use AI on my own projects, the code it generates basically looks like auto-complete to me; exactly what I would have written by hand. So I don't see why it needs to be marked as AI. It's just saving me time. It's very much my own work.

If your code style involves duplicating code, then LLMs will duplicate a lot of code. If you reuse code, LLMs will reuse code.

If your code is full of patches and workarounds, then LLMs will write code that is full of patches and workarounds.

That's the inference aspect.

There are also some biases introduced by the training set itself. If your favorite programming language's training set favors over-engineering and an enterprise coding style, then LLMs will leans towards that coding style.

This is why I prefer Javascript over Typescript. TS is more enterprisey. In TS, LLMs tend to invent unnecessary abstractions and create architectural complexity... TS was designed for large corporate teams of people who move slowly, bill by the hour and whose salary and job security is determined primarily by their lock-in factor within the company; their incentives aren't necessarily aligned with delivering results and this is reflected in LLM output.

Though I agree with OP's point of view that MCP was overhyped for too many use cases. Complex agent-system integration problems are usually better solved with just AI agent + curl + SKILL.md. It's just way more flexible.

It's another variant of the 'fat client/thin server vs thin client/fat server' debate. Some people want rigid, thin (e.g. web-based) frontends with the LLM doing work in secret behind the scenes. Others want fat, versatile frontends through which the LLM can interact with the user's own environment.

I've always been a fat client guy and this time is no exception. I doubt the constrained approach is going to lead ground-breaking innovation. I also wish companies would treat SKILL.md + curl as the main mechanism for agent tool calling as opposed to MCP. MCP is niche.

The people who prided themselves on raw puzzle-solving ability got hit hard. Those who are idea-driven and architecture-oriented feel like they got handed a superpower.

It's quite a big shock at the industry level because the entire software engineering job interview process at essentially all large companies was heavily biased towards well-defined, raw problem-solving under time constraints which is precisely the skill which AI has replaced.

Now AI has surpassed me in fine-grained problem-solving ability. It can write more reliable code than I can, faster than I can, provided it is given the right guidance.

But one thing that I'm still much better at is identifying technical opportunities and choosing the right tradeoffs.

My feeling is that frontier models are incredibly smart in some ways, but incredibly dumb in other ways

When I chat with Claude using deep technical language about distributed systems issues, the arguments it presents are mind-blowingly good. I worked with many skilled engineers on complex projects but the kinds of arguments Claude makes are on another level. It feels like it can read my mind because it already identified all of the relevant aspects to the current topic and it's incredibly persuasive. I would say it's even better at deep, nuanced technical discussions than I am.

When I ask it to implement a feature using technical language, it does a really good job and the solution usually works out of the box. No bugs at all 95% of the time even after adding 1000 lines!

But the problem it still has is that when it implements solutions, it misses so many low-hanging fruits/opportunities. Especially in terms of performance, maintainability, scalability and UX. It's like it sees and recalls all the relevant parts perfectly but yet somehow misses opportunities which seem extremely obvious to me.

It feels like AI has 0 creativity. It only sees the opportunity once I mention it... And once it sees the opportunity, it demonstrates deep understanding of the technical implications. I think what's surprising is that it understands the suggested solution so well, with such nuance, that I can't understand how it didn't see the opportunity and why it never seems to see it until I mention it.

This is very unhuman-like. There is no way that a human being with that degree of understanding of a topic would be presented with such highly relevant context and not make the connection.

That said, I also think everyone who contributed content online deserves to be rewarded for their contribution, especially since OpenAI (which made the largest contribution) was a non-profit in the beginning. So there should be some kind of tax. Otherwise I agree, it's not ethical. It's not 'fair use' of copyrights.

It's going to destroy a lot of startups which were monetizing this exact idea. But clearly it's a low-hanging fruit so it makes sense that OpenAI would do it.

Seems we're on the same wavelength because I had the exact same thought about this.

But actually, I'm probably not so fussy. I wouldn't mind cleaning up the mess if I'm paid hourly... But now my concern is: What if they want me to clean up the mess but demand that I do it in a particular way which makes it impossible?

I'm already seeing signs of this. This was already kind of the case in my last job; I had to fix things and implement new features but we had to keep the same clunky, over-engineered architecture. My current job doesn't have the same degree of architectural legacy baggage but there is a lot of bureaucracy to deal with instead, which is itself restrictive.

Just watch the software engineers who say "Let's be careful" lose their jobs and get replaced by non-technical vibe-coders churning out 10K lines of dirty insecure, unmaintainable code per day... Ticking time bomb.

I swear, good engineers are going to move to North Korea to monetize because of the amount and size of 'opportunities' this will create.

It seems so far-fetched but that's the direction it seems to be heading.

There are project I've built from scratch that I would feel confident to hand off to a bunch of non-technical vibe coders and I know they would be productive and the product would likely be secure; because the existing codebase already exhibits all the patterns and principles that are required for that kind of project.

It would probably slowly degrade over time if a lot of vibe-coded logic is added on top but I think they could get very far feature-wise whilst keeping the software reliable.

But even though the value of such codebase has increased, people haven't adapted to this new reality. People are generally not good at telling what is good code. Because we don't actually have consensus on a definition. My definition is that good code is code that is easy to extend and maintain.

If implementing a feature requires a huge amount of tokens, then there's a good chance the codebase is not great.

I've worked on a codebase where a small feature requires might require 3k tokens, but on a different codebase, a feature of similar complexity would require 30k tokens minimum... And it's not about the size of the project; it's more about how the logic is divided and the architecture. And importantly; it's not a one-off; it's a clear observable, repeatable pattern.

Anyway databases nowadays are a commodity. A sticky commodity but nonetheless they are replaceable; increasingly so in the age of AI where data migrations are easier than ever.

Even if each user gets their own sandbox, they will still want to configure different access rules for different kinds of data which they host.

That said the idea that each user could control and host their own data is interesting and could work. I imagine you could have apps which link data from many different user sandboxes via remote foreign keys.

You could have a centralized data schema controlled by the application owner but the data itself would be held/scattered across a large number of sandboxes.